Turnbull v Warrener

Nadine Turnbull and Rebecca Warrener are experienced horsewomen. Mrs Warrener was the owner of a horse called Gem. Between 2001 and July 2005 Mrs Warrener regularly rode Gem. She ceased to do so only when she became pregnant. In November 2005 she made an arrangement with Ms Turnbull that Ms Turnbull would ride Gem on a regular basis. This would enable Gem to continue to exercise and would enable Ms Turnbull to have some extra riding. For about four months Ms Turnbull rode Gem most weekends. On 26 March 2006 Ms Turnbull was riding Gem when, in circumstances I shall later describe in more detail, she fell and sustained injuries. Until that day, Gem had never disobeyed his riders’ instructions or failed to respond to his riders’ attempts to control him. Ms Turnbull commenced proceedings against Mrs Warrener in the Oxford County Court alleging negligence and/or liability pursuant to section 2(2) of the Animals Act 1971. She failed under both heads. As regards negligence, her claim fell apart when the judge rejected her evidence as untruthful in relation to important disputed matters. The statutory cause of action failed because the judge held that Ms Turnbull was unable to satisfy the conditions set out in section 2(2) and that, in any event, Mrs Warrener would have been entitled to succeed on the basis of the statutory defence pursuant to section 5(1) on the basis that the damage was “due wholly to the fault of the person suffering it”.

On this appeal, Ms Turnbull does not seek to resurrect her negligence action. Her case is that the conditions prescribed by section 2(2) were satisfied and that the statutory defence pursuant to section 5(1) did not arise. For her part, Mrs Warrener maintains that the judge was correct about section 2(2) and section 5(1). In addition, by a Respondent’s Notice, she maintains that she was also entitled to resist the claim by reference to the defence of voluntary acceptance of risk pursuant to section 5(2).

Section 2 of the 1971 Act is headed “Liability for damage done by dangerous animals”. It distinguishes between animals which belong to a dangerous species and those which do not. As regards the former, section 2(1) provides:

“Where any damage is caused by an animal which belongs to a dangerous species, any person who is a keeper of the animal is liable for the damage, except as otherwise provided by this Act.”

That is undoubtedly a form of strict, albeit not absolute, liability. Horses are not a dangerous species.

If section 2(1) can be said to be drafted with commendable simplicity, the same cannot be said for section 2(2). The drafting, which has attracted four decades of judicial and academic criticism, is grotesque. The leading authority is Mirvahedy v Henley [2003] 2 AC 491. Even before that case had reached the House of Lords, in the Court of Appeal Hale LJ gathered together the numerous expressions of judicial disapprobation to which it had given rise: [2002] QB 769, at paragraph 18. Section 2(2) provides:

“Where damage is caused by an animal which does not belong to a dangerous species, a keeper of the animal is liable for the damage, except as otherwise provided by this act, if –

(a)

the damage is of a kind which the animal, unless restrained, was likely to cause or which, if caused by the animal, was likely to be severe; and

(b)

the likelihood of the damage or of its being severe was due to characteristics of the animal which are not normally found in animals of the same species or are not normally so found except at particular times or in particular circumstances; and

(c)

those characteristics were known to that keeper or were at any time known to a person who at that time had charge of the animal as that keeper’s servant or, where that keeper is the head of a household, were known to another keeper of the animal who is a member of that household and under the age of sixteen.”

Plainly liability for an animal which does not belong to a dangerous species was intended to be less strict than that in relation to an animal which belongs to a dangerous species. In order to establish liability under section 2(2), a claimant has to satisfy all three of the prescribed conditions.

Section 5 provides exceptions from liability under section 2 and other provisions. For present purposes I need refer to only two of the exceptions:

1)

“A person is not liable under sections 2 – 4 of this Act for any damage which is due wholly to the fault of the person suffering it.

2)

A person is not liable under section 2 of this Act for any damage suffered by a person who has voluntarily accepted the risk thereof.”

Thus, for the purposes of section 2, these statutory defences are available whether or not a particular animal belongs to a dangerous species.

As I have said, Ms Turnbull began to ride Gem in November 2005 and did so regularly until the accident on 26 March 2006. For some weeks prior to the accident, there had been a problem with Gem in that he had tended to pull his head to the right. It was thought that this might be a result of his having a sore mouth and so, a few days before the accident, Gem was taken to the dentist where he had his teeth filed. The dentist advised Miss Warrener that Gem should not be ridden using a bridle with a bit for the following week. As a result Mrs Warrener borrowed a bitless bridle from a friend. On 26 March Ms Turnbull had a discussion with Mrs Warrener about riding Gem with the bit-less bridle. Gem had never experienced a bitless bridle before but Ms Turnbull was content to ride. The two women tacked up Gem together and fitted the bitless bridle. As equally experienced horsewomen, they both knew that it would be prudent for riding to commence in confined conditions. Ms Turnbull rode Gem for about five minutes in the lungeing school before walking and trotting him in circles for 10 – 15 minutes in another enclosed area. They then moved to an outdoor space for 15 -20 minutes. At no stage up to this point did Ms Turnbull indicate that she was having any difficulty with Gem and Mrs Warrener did not observe Gem reacting differently to the bit-less bridle. After Ms Turnbull had ridden Gem in the outdoor space, Ms Warrener suggested that they call it a day. Ms Turnbull replied that she wanted to take Gem for a canter up the field. Ms Warrener said that if Ms Turnbull felt safe cantering up the field, she could do so. When Ms Turnbull set off Gem appeared to be going more quickly than had been expected – more like a gallop than a canter – and Ms Turnbull was visibly trying to restrain him. However, Gem suddenly veered to the right and went through a gap in a hedge. At this point Ms Turnbull fell off, landing on a tarmac area and sustaining her injuries.

The trial took place over two days in Oxford County Court. There were serious factual disputes which, as I have said, were resolved in favour of Mrs Warrener. Experts were called on behalf of both parties: Mr Charlie Lane on behalf of Ms Turnbull and Mr Richard Meade OBE on behalf of Mrs Warrener. Because the disputed evidence took as long as it did, there was no time for oral submissions about the complexities of section 2(2). Having dismissed the claim of negligence, the judge dealt with the statutory cause of action in fairly short order. He said:

“There is no evidence that Gem had ever before behaved in the way he did immediately prior to C’s accident, namely by refusing to respond to her attempts to control him. I would find that he was not likely to cause injury unless restrained. He was (and probably still is) an ordinary domesticated horse. He was described by D as being quite lazy and needing ‘some encouragement’ before he would canter. I would find further that any injury caused by him was not likely to be severe. I base such a finding on the evidence of Mr Lane. I find that he had no abnormal characteristics. There is no evidence that he did. Was he displaying a characteristic normally found in horses at particular times or in particular circumstances? What was the characteristic that he was displaying? … I find that Gem did not ‘bolt’ in the technical sense of the word, ie he did not take off headlong out of control without regard for his own safety or that of C. Rather, he was going faster than she wanted him to go and she was unable to stop him. I accept Mr Lane’s evidence (and I do not think Mr Meade disagreed with this) that the reason why she was unable to stop him was that she was trying to do it by means of a piece of equipment, namely the bitless bridle, to which Gem was unaccustomed. It was not a case of Gem going off faster than C meant him to go because he was fitted with the bitless bridle. Was it ‘normal’ for horses not to respond to the rider’s instructions in these circumstances? I can find no evidence to support such a contention. If that be wrong and there is evidence to support it, there is no evidence to support the proposition that D knew of such a ‘characteristic’, whether in Gem or in horses generally. Gem had never before disobeyed his rider’s instructions and/or failed to respond to his rider’s attempts to control him. Furthermore, I find it difficult to see that Parliament intended that D should be strictly liable to C for the consequences of this accident on the facts of the case as I have found them to be. Finally, if all the reasoning above with regard to the claim under the Animals Act is wrong, I would hold that the injury that C suffered was wholly due to her own fault in cantering off on Gem as she did using a bitless bridle before testing him adequately with that piece of equipment at that speed in closed and/or open conditions. In those circumstances D is not liable for her injury by reason of the provisions of section 5(1) of the Act. I would not find that C had voluntarily accepted the risk she took in doing what she did that day for the purposes of section 5(2).”

The evidence of Mr Lane in relation to the likelihood of severe injury which was referred to by the judge was contained in this passage of cross-examination:

“Q: Would you agree that people fall off horses on hundreds of occasions every day up and down the country?

A: At least hundreds every day.

Q: And very few of those falls result in severe injury?

A: Certainly a low percentage.

Q: The majority result in no injury or very minor soft tissue injury?

A: Yes.

Q: So a fall in the circumstances such as this may possibly result in severe injury but its not likely?

A: It might occur, but statistically it probably won’t.”

The experts had set out the extent of their agreement and disagreement on relevant issues. They were in agreement as follows:

“We agree that both Claimant and Defendant should have known that all horses can be unpredictable on occasions and that such unpredictable behaviour can include running away/bolting and that riding is a risk sport and the risk is that the rider may be unbalanced and fall or in some other way receive injury …

We agree that the fitting of a bitless bridle to allow Gem to be ridden while his sore mouth healed was a sensible decision …

We agree that the only way to tell whether a horse will go well in a bitless bridle is by trial. It will be important that the horse’s first ride is in the relatively controlled environment of a riding arena or some such enclosed area in case the horse is difficult to control.

We agree that quite how long that the horse should be ridden in the enclosed area must be a matter of judgment but the horse must show that he will slow down and stop from walk, trot and canter. He should then be ridden outside of the area on routes and in areas with which he is familiar and show that he will slow down and stop from walk, trot and canter. When cantering is first practised this should be in a place and direction where the horse will not be inclined to charge off, for instance going away from home or in an area where the horse is familiar with being ridden in that area.

We agree that it is common sense that the horse is taken progressively through each stage and that the next stage is only moved on to when the rider is confident that the horse is under control. It is not possible to stipulate a length of time for each stage. The length of time for each stage must be a matter of judgment.

We agree that the claimant as the rider should have been in a better position than the Defendant to assess how Gem was responding to being ridden in the bitless bridle.”

The experts produced a manual called The Bitless Bridle by Dr Robert Cook. It contained these words of caution:

“When first introduced to the bitless bridle, it sometimes revives a horse’s spirits with a feeling of ‘free at last’. Such a display of exuberance will eventually pass but be prepared for the possibility even though it occurs in less than 1% of horses. Begin in a covered school or small paddock rather than an open area. Consider preliminary lungeing or a short workout in the horse’s normal tack. These and other strategies familiar to horse people can be used to reduce the small risk of boisterous behaviour.”

I now turn to the difficult issues which arise in relation to section 2(2).

There are two limbs to section 2(2)(a). In order to succeed, a claimant has to satisfy at least one of them. In interrogatory form, they are: Is the damage of a kind which the animal, unless restrained, was likely to cause? Alternatively, is the damage of a kind which, if caused by the animal, was likely to be severe? As to the former, the judge concluded that, because Gem had never behaved in such a way before and he was, and probably still is, an ordinary domesticated horse, he was not likely to cause the kind of damage in question. That finding is not challenged. However, there is an issue in relation to the second limb. The judge concluded that Ms Turner had failed to satisfy it. He relied on the evidence of Mr Lane which I have set out in paragraph 9, above.

This is novel. In previous cases, the point seems to have been conceded by the defendant or readily assumed by the judge in favour of the claimant: see Mirvahedy, per Lord Nicholls at paragraph 15 (but cf Lord Scott, dissenting, at paragraphs 95-98); Welsh v Stokes [2007] EWCA Civ 796, per Dyson LJ at paragraph 33; Freeman v Higher Park Farm [2008] EWCA Civ 1185, per Etherton LJ, at paragraph 34; Clark v Bowlt [2006] EWCA Civ 978, per Lord Phillips CJ, at paragraph 11; and Goldsmith v Patchcott, per Jackson LJ, at paragraph 33:

“It should be noted that this subsection will only eliminate a small number of cases. Most animal-related damage which someone wishes to sue about will fall into one or other of those two categories.”

The question for us is whether in the present case the judge was entitled to come to the contrary conclusion by reference to answers given in the course of cross-examination by the claimant’s expert.

Although there have been semantic debates about the meaning of “likely” in section 2(2)(a), it seems to me that it has now become settled. It means “reasonably to be expected”: see Mirvahedy, per Lord Scott, at paragraphs 95-97, and Freeman, per Etherton LJ, at paragraph 33.

Freeman is perhaps the most interesting authority on this issue. It had been submitted on behalf of the horse owner that it was incumbent upon the claimant rider to adduce evidence that physical injury, if caused by the horse, was likely to be severe. The suggestion was that such evidence should be of a statistical nature, quantifying accidents and correlating them to serious injuries. Etherton LJ rejected this as “quite unrealistic and unnecessary”, adding (at paragraph 34):

“It is obvious that, if a horse bucks on beginning to canter so that the rider falls off, it is reasonably to be expected that severe injury will result.”

He drew on Welsh, where Dyson LJ had treated the preposition as “self-evident”.

In the light of this approach, one cannot blame a claimant for failing to adduce expert evidence of the likelihood of severe injury. What, then, is the position if the claimant’s expert gives answers in cross-examination which are helpful to the defendant on this issue? Is it open to the judge to displace what was previously considered to be obvious? Or was it impermissible (that is, perverse) so to conclude in this case?

The factual circumstances of each case inevitably differ. Of the four material questions asked in cross-examination, the first three were general in nature and elicited answers that strike me as controversial. The majority of falls from horses do not result in serious injury. It was the fourth question and answer that influenced the conclusion:

“Q: So a fall in circumstances such as this may possibly result in severe injury but it’s not likely?

A: It might occur, but statistically it probably won’t.”

The question is whether that, in itself, justified the conclusion that the second limb of section 2(2)(a) was not satisfied.

Notwithstanding the skill of the cross-examination, I do not consider that its product was sufficient to produce the heterodox finding which eventuated. At most, it discounted a probability. However, that is not the same as a likelihood, with its received meaning of “reasonably to be expected”. Moreover, in the circumstances in which the judge was considering the issue (hurriedly, and without the benefit of oral submissions late on the second day of the trial), it is not clear that he had in mind the “reasonably to be expected” test. In my judgment, the decision on section 2(2)(a) was not a permissible one, either because it was not really supported by the evidence and was contrary to received wisdom or because it resulted from a failure to apply the correct test.

This is the provision which has attracted the most criticism for being “opaque”, “tortuous” or, most recently, “oracular” (per Jackson LJ in Goldsmith, at paragraph 31). I am bound to say that, if we were considering its meaning and application untroubled by previous authority, I would hope to be able to uphold the judge’s finding because I do not think it was the intention of Parliament that circumstances such as those in the present case should give rise to liability under the Act. In his dissenting speech in Mirvahedy, Lord Scott said (at paragraph 130):

“It seems to me that Parliament was trying to draw a distinction between animals that in normal circumstances behaving normally are dangerous and those that in normal circumstances behaving normally are not. As to the former, they belong to a dangerous species and there was to be strict liability for damage [pursuant to section 2(1)]; as to the latter they do not belong to a dangerous species and strict liability was to be limited to damage caused by the animal displaying abnormal characteristics that it was known by its keeper to possess. This seems to me to be a coherent policy. In respect of damage for which no strict liability was imposed, a remedy in negligence would always be available if the keeper of the animal had failed to exercise reasonable care to see that the animal did not cause damage.”

Elsewhere, (paragraph 103) Lord Scott opined that, in relation to section 2(2) animals: “the keeper’s primary liability for damage would be a liability in negligence”.

I find Lord Scott’s approach compelling but his view did not prevail. The approach by which we are bound is very favourable to a claimant.

Lord Nicholls expressed the majority view in Mirvahedy succinctly (at paragraph 47):

“The fact that an animal’s behaviour, although not normal behaviour for animals of that species, was nevertheless normal behaviour for the species in the particular circumstances does not take the case outside section 2(2)(b).”

In an earlier passage, he had acknowledged that this interpretation “seems to leave section 2(2)(b) with very limited content” (paragraph 42). In Goldsmith, Jackson LJ went further, saying “it is not obvious … what purpose section 2(2)(b) serves” (paragraph 40). The problems for the defendant are compounded by the fact that the word “normally” which appears twice in section 2(2)(b) does not carry the meaning “usually” but the less frequent “conforming to type”: Welsh, per Dyson LJ, at paragraphs 44-46.

None of this subtlety featured in the oral submissions to the judge in the present case. In considering section 2(2)(b), he asked himself the question: “Was it ‘normal’ for horses not to respond to the rider’s instructions in these circumstances?” He would find “no evidence to support such a contention”. It seems to me that this approach was too simplistic, having regard to the authorities by which we are bound. The judge accepted that “the reason why [Ms Turnbull] was unable to stop him was that she was trying to do it by means of a piece of equipment, namely the bitless bridle, to which Gem was unaccustomed”. This ought to have led to consideration of whether refusing to respond to instructions given through the bitless bridle was a characteristic of horses unfamiliar with such equipment. In my judgment, such consideration would or should have resulted in the answer that it was such a characteristic. That was the whole point of the initial confinement to the enclosed area. In his most recent written submission, Mr David Regan puts it as follows:

“… by refusing to slow down on command or at all when using new equipment, specifically the bitless bridle, Gem was displaying a characteristic normal to horses in that he was conforming to type and displaying a natural characteristic, even if one unusually occurring. It was an effect of the equipment which the users’ manual specifically warned against.”

I find it difficult to reject this submission. At the very least, it raises proper questions which the judge did not address. This lends me to the conclusion that the judge’s finding that section 2(2)(b) was not satisfied cannot stand.

This conclusion is shrouded in reluctance. The drafting of section 2(2)(b) has resulted in its virtual emasculation. It will become apparent that Stanley Burnton and Lewison LJJ share my concern. I refer again to the speech of Lord Scott (with whose dissent Lord Slynn agreed). He said (at paragraph 118):

“If there was intended to be strict liability for damage caused by an animal behaving in a manner normal for its species, one would have expected the Act to categorise the species as a dangerous species. The fact that an animal belongs to a species that falls outside the statutory definition of a dangerous species is, in my opinion, an indication that behaviour by the animal in a manner normal for the species was not intended by Parliament to attract strict liability. If there was intended to be strict liability for damage caused by normal behaviour of non-dangerous animals one would have expected that simple proposition to be simply stated rather than left to be produced by the literary device of turning the double negative in the concluding words of paragraph (b) into a positive.”

See also Sedley LJ in Clark, at paragraph 24.

In March 2009, DEFRA produced a document entitled Consultation on changes to the Animals Act 1971 to clarify the application of strict liability to the keepers of animals. It contained modest proposals which would breathe life into section 2(2)(b). So far it has not borne statutory fruit.

The judge held that “there is no evidence to support the proposition that [Ms Warrener] knew of such a ‘characteristic’, whether in Gem or in horses generally”. This, too, was a questionable conclusion. The law was expounded by Dyson LJ in Welsh (at paragraph 71):

“I do not see why a keeper’s knowledge that a horse has the characteristic of normally behaving in a certain way in particular circumstances cannot be established by showing that the keeper knows that horses as a species normally behave in that way in those circumstances … It makes no sense to require a keeper, if aware of that general characteristic, to have some additional and more particular knowledge.”

Ms Warrener knew that all new tack should be tried out in an enclosed environment first because one never knows how a horse may react; that this is a necessary precaution; and that that was why she was present and ensured the initial use of a closed environment. She therefore knew that horses may react unpredictably on using equipment with which they are unfamiliar. The expert evidence was that, because of the risks, horses should progressively be taken through walking, trotting and cantering in an enclosed environment and that the reasonably competent rider such as Mrs Warrener should know that the horse might react unpredictably. All this provided an evidential foundation for a finding of knowledge. The judge’s reliance on the fact that Gem had never before disobeyed his rider’s instructions or failed to respond to his rider’s attempts to control him ought not to have been determinative.

It follows from what I have said that, in my judgment, the conclusions of the judge on section 2(2) were flawed. In these circumstances, I turn to the statutory defences.

The keeper of an animal is not liable under section 2 for any damage “which is due wholly to the fault of the person suffering it”. The judge considered that, if the requirements of section 2(2) had been satisfied, Ms Turnbull would still have failed by reason of this defence. “Fault” in this context is given the same meaning by section 11 as in section 4 of the Law Reform (Contributory Negligence) Act 1945, viz:

“negligence, breach of statutory duty or act or omission which gives rise to a liability in tort or would, apart from this Act, give rise to the defence of contributory negligence.”

There is an obvious difficulty with the judge’s finding that the damage was “wholly due to [Ms Turnbull’s] fault in cantering off on Gem as she did using a bitless bridle before testing him adequately with that piece of equipment in closed and/or open conditions”. Ms Turnbull and Mrs Warrener were horsewomen as between whom there was “no material distinction to be drawn” in relation to their respective riding experiences and ability. On the judge’s finding, it was not negligent of Mrs Warrener to permit Ms Turnbull to proceed to canter at the point when she did. The expert witnesses agreed that Gem should have been cantered first in the enclosed area. Mrs Warrener could have insisted on that but she did not. In these circumstances, to find that Ms Turnbull was “wholly” at fault cannot coexist with the finding that Mrs Warrener was not negligent. In my judgment, the defence under section 5(1) could not avail Mrs Warrener. Either both or neither were at fault in the statutory sense.

By a respondent’s notice, Mrs Warrener challenges the judge’s conclusion that she could not establish the defence under section 5(2), that is that Ms Turnbull had “voluntarily accepted the risk”. The judge did not explain that conclusion. The law on section 5(2) was explained by Etherton LJ in Freeman (at paragraph 48):

“The words of section 5(2) are simple English and must be given their ordinary meaning and not be complicated by fine distinctions or by reference to the old common law doctrine of volenti … what must be proved in order to show that somebody has voluntarily accepted the risk is that (1) they fully appreciated the risk, and (2) they exposed themselves to it.”

There are factual similarities between Goldsmith and the present case. The horse there was normally docile but both claimant and defendant knew that “horses could buck when startled or alarmed”. Neither of them anticipated that the horse would buck as violently as it did on the day of the accident. Nevertheless the section 5(2) defence succeeded. This was on the basis of knowledge of the inherent characteristics of horses in general coupled with choosing to ride notwithstanding knowledge of the risks consequent upon such inherent characteristics. Moreover, Mrs Warrener, who had suggested terminating the ride, had told Ms Turnbull that she could canter Gem up the field if she felt it would be safe to do so.

In the present case, Ms Turnbull knew that a horse, just fitted with a bitless bridle for the first time, bore an increased risk of not being responsive to a rider’s instructions. That was the whole point of the initially cautious approach in the enclosed area. She also knew that, when she took Gem into the open in order to canter, he had not yet cantered when fitted with the bitless bridle. In these circumstances, it is plain that she had voluntarily accepted the risk which eventuated. It is a fact that, as Jackson LJ said in Goldsmith (at paragraph 40), horses “may act out of instinct or impulse and their precise behaviour cannot necessarily be predicted”. He also cited with approval the judgment of Mr Recorder Ryan in Plum v Berry in Preston County court on 5 November 2004:

“… the risk of being unseated by such an event is one which any rider who is competent to canter and hack accepts. Riding is a pursuit involving the control by a rider with the wind and the aids of rein, leg, seat and crop of a horse with its own mind and physical attributes. That relationship and activity involving two living beings cannot be precisely predicted or judged to the second or the centimetre. The occurrence of an accident in such a manner as I have found is precisely the risk and type of risk which a rider undertakes.”

In my judgment that has particular resonance in the present case. Of course it does not mean that every injured rider of competence will be defeated by section 5(2). Flack v Hudson [2001] QB 698 is an example of section 5(2) not availing the horse owner – she knew of the specific risk-creating characteristic (fear of agricultural machinery) but the rider, who received fatal injuries, did not.

There is a further point which underwrites the section 5(2) defence in the present case. It arises from the equivalence of knowledge and experience as between the parties. If Mrs Warrener’s knowledge for the purpose of section 2(2)(c) is established, it is difficult to see how knowledge as an element of voluntariness on the part of Ms Turnbull for the purpose of section 5(2) can be denied. There is, on the judge’s undisputed findings, a significant symmetry in this case.

For all these reasons, I consider that Mrs Warrener has the defence provided by section 5(2).

This is an unusual case. Having disagreed with the judge about almost everything relating to statutory liability, for the reasons I have given I have come to the same ultimate conclusion. Ms Turnbull’s claim fails and her appeal should be dismissed.

Like Maurice Kay LJ and Lewison LJ (whose judgment I read in draft) I am troubled by the interpretation of section 2 of the Act in the authorities to which we have been referred and by which we are bound. The result of these authorities, and in particular the judgments of the majority of the House of Lords in Mirvahedy, hitherto has been that there is no significant difference between the liability of the keeper of a wild animal and that of a keeper of a domesticated animal. I have no doubt that that was not the intention of Parliament in enacting the Act. I particularly wish to associate myself with paragraphs 2 to 7 of the judgment of Lewison LJ.

I agree that this appeal must fail. In the light of my Lords’ judgments, I can express my views quite shortly.

First, in my judgment it was open to the judge to find, on the evidence before him, that the requirement in section 2(2)(a) was not satisfied. I agree with paragraphs 8 to 13 of the judgment of Lewison LJ. The judge was bound to consider the application of that paragraph on the evidence before him, rather than on what judges in previous cases had considered to be obvious as a matter of fact. The effect of Mr Lane’s evidence was not wholly clear, but neither side sought to clarify it, and it was left to the judge to assess its effect. It was open to him to make the finding he did.

Secondly, I question whether section 2(2)(b) was satisfied. According to The Bitless Bridle, to which Maurice Kay LJ refers at paragraph 9 of his judgment, exuberance on the part of a horse fitted for the first time with a bitless bridle occurs in less than 1 per cent of horses. I take that to have been accepted by both parties’ experts. A characteristic of an animal is something inherent in the animal. An unprecedented, one-off action of an animal is not a characteristic of that animal. It is difficult to say of action that occurs in less than 1 per cent of horses fitted with bitless bridles that it is characteristic of them at particular times or in particular circumstances.

Lastly, I entirely agree, for the reasons given by Maurice Kay LJ, that the defence in section 5(2) succeeds, and that in section 5(1) did not.

I agree that the appeal should be dismissed although I have reached my conclusion by a different route.

I cannot help but express my concern about the way that the law has developed. It is well known that the Animals Act 1971 finds its genesis in the Law Commission’s report on Civil Liability for Animals. The purpose of the report was to make recommendations for the modernising and simplifying of the common law. The Commission noted that at common law there was strict liability for damage caused by animals belonging to a dangerous species. They proposed to retain that rule. They went on to discuss damage caused by animals belonging to non-dangerous species. In paragraph 17 they said:

“If, as we have recommended, there is to be a category of animals of a dangerous species for which strict liability is imposed, it would seem reasonable that an animal not belonging to that category should nevertheless give rise to strict liability in respect of injury or damage which it causes if that damage results from dangerous characteristics of the particular animal which are known to its keeper. As far as the potential defendant is concerned, he is equally the creator of a special risk if he knowingly keeps, for example, a savage Alsatian as if he keeps a tiger. As far as the potential plaintiff is concerned, an animal belonging to an ordinarily harmless species, which is known to its keeper to be dangerous is in the nature of a trap—a "wolf in sheep's clothing"—which would seem to justify the same strictness of liability as applies to an obviously dangerous animal.”

Accordingly in paragraph 18 they said:

“The law at present achieves this imposition of strict liability by the scienter rule, but we think that this rule requires considerable modification and simplification. We would therefore abolish it in its common law form and substitute a new rule retaining what we conceive to be the essential rationale of the old law.”

The essential rationale of the old law, from which the Law Commission did not intend to depart, was that liability attached to animals which had a propensity to attack, either because they belonged to a dangerous species, or because the particular animal had that propensity. The latter propensity was variously described as “vicious, mischievous or fierce”: see Fitzgerald v ED and AD Cooke Bourne (Farms) Ltd [1964] 1 QB 249. Thus the cases concerned the dog that bit, the horse that kicked, the ram that butted, the bull that charged, and so on. As the Law Commission themselves pointed out (§ 6): “It would seem that the act of the animal must be in the nature of an “attack” and does not therefore include behaviour which, although it may cause damage, is merely frolicsome.”

You will search the classic pre-Animals Act text book (Glanville Williams: Liability for Animals (1939)) in vain to find a case in which a horseman or horsewoman who lost control of a horse and fell off successfully sued the owner of the horse for damages. I suspect that our Georgian and Victorian ancestors, for whom the horse was a common means of transport, would have scoffed at the idea.

As I have said, the Law Commission did not proclaim an intention to widen the existing scope of the law to the extent that would be necessary to catch an ordinary riding accident. Moreover, as Lord Nicholls pointed out in Mirvahedy v Henley [2003] UKHL 16 [2003] 2 AC 491 (§ 39) when the Law Commission’s draft bill was introduced into Parliament the criticism was that the net of strict liability was being cast too widely. So the bill was redrafted to confine the scope of section 2. In the same case Lord Walker said of section 2 (§ 135):

“In practice section 2(1) has a very narrow scope, being almost entirely limited to incidents in (or following escapes from) zoos or circuses. Section 2(2) has to cover the whole range of incidents involving animals of species classified as non-dangerous (which I will call domesticated animals, although that is not an entirely accurate term). That range includes (i) physical injury to humans by biting (especially by dogs) or kicking or knocking down (especially by horses); (ii) injuries caused to livestock (such as a dog worrying a neighbour's sheep, or a cat killing a neighbour's chickens); (iii) road traffic accidents, especially those caused by animals straying on the highway; (iv) damage caused by livestock getting out on to neighbouring land and destroying crops or gardens; and (v) injury or damage caused by the spread of animal infection or by the smell or noise of animals (a class which shades off into cases normally classified as nuisance). So section 2(2) has a lot of work to do. It is expressed in general, abstract terms and it has to be applied to a wide range of disparate incidents.”

There is no hint here that an ordinary riding accident would be covered. Mirvahedy was itself a case of a horse colliding with a car. The two alternative interpretations of section 2 (2) before the House were laid down in cases concerning respectively a dog that attacked (Cummings v Granger [1977] 1 QB 397) and a horse that kicked (Breeden v Lampard (21 March 1985). The damage suffered in these two cases would also have been within the general scope of the old scienter action.

In the riding accident cases which have been drawn to our attention (Welsh v Stokes [2007] EWCA 796 [2008] 1 WLR 1224 and Freeman v Higher Park Farm [2008] EWCA Civ 1185) it has been assumed that personal injury resulting from a rider losing control of a horse and falling off is “damage caused by” the horse. The same assumption was made in the present case. The assumption may be right; but it has never, so far as I am aware, been tested by argument. If it is right, then the Animals Act 1971 has caused a major (and unintended) expansion in the potential scope of strict liability.

There is also a difficulty in the way that the damage is characterised for the purposes of section 2 (2)(a). In Welsh v Stokes the trial judge, Judge Tyzack QC, started with the actual damage caused. He said:

“The starting point here is the accident itself. In other words, Ivor rearing up and causing the claimant to fall off onto the road. In my judgment, this requirement is satisfied here, because personal injury arising from such an accident is likely to be severe. Anyone falling off a horse that has reared up and falling onto a tarmac road is likely, in my judgment, to suffer severe injury; still more so is this likely to be the case, in my judgment, if the horse falls backwards onto the rider. Thus, I find the second part of requirement (a) is satisfied.”

This finding was not challenged on appeal (§ 33, 40). If you start with severe injuries resulting from the accident itself and the particular circumstances in which it took place, then the question whether the damage is “likely to be severe” answers itself. If the injury was in fact severe, it will be almost inevitable that, barring some freak, it was likely to be severe. To borrow from Lord Hobhouse in a case of occupier’s liability (Tomlinson v Congleton BC [2004] 1 AC 46 § 79): “It is a fallacy to say that because drowning is a serious matter there is therefore a serious risk of drowning.” The problem, as I see it, is the level of generality at which you ask and answer the question. If the question is: if you fall off a rearing horse onto a hard surface, and the horse falls on top of you, is the injury likely to be severe, you may get one answer. But if the question is: if you fall off a horse is the injury likely to be severe, you may get another. I do not believe that this problem has yet been squarely confronted in the cases.

In the course of his judgment in Welsh v Stokes Dyson LJ said (§ 40):

“If a horse rears in the particular circumstances and the rider falls from the horse, she is likely to suffer severe damage. That may be because she falls on a hard surface (which need not be a tarmac road) or because the horse falls on her or in some other way. I do not understand Miss Rodway to challenge this.” (Emphasis added)

It is clear that the starting point for the enquiry was not challenged. In Freeman v Higher Park Farm Ms Freeman fell off a horse called Patty which bucked when going into a canter. The Court of Appeal rejected the argument that in determining whether the injury caused by Patty was likely to be severe there should have been general evidence of injuries throughout the country sustained by riders who fell of horses. Etherton LJ said (§ 34):

“That is quite unrealistic and unnecessary. The evidence was that, if physical injury was caused by Patty, it would have been caused when she bucked when beginning to canter. There was no evidence of the likelihood of physical injury at any other time or in any other circumstances. It is obvious that, if a horse bucks on beginning to canter so that the rider falls off, it is reasonably to be expected that severe injury will result. In Welsh v Stokes [2007] EWCA Civ. 796 at para [40] Dyson LJ, with whom the other members of the Court of Appeal agreed, regarded this as self-evident in the case of a rider falling from a horse who rears. I see no reason why the same approach should not be appropriate in the case of a rider falling from a bucking horse about to canter.”

These statements seem to me to be statements of fact rather than rulings on the law. In the present case the judge heard expert evidence on the likelihood of serious injury, which Maurice Kay LJ has quoted. The evidence was that riders fall off horses every day and do not sustain severe injury. I would not characterise the first three questions as controversial. Almost anyone who has ever ridden will have the experience of having fallen off a horse, getting up and remounting the horse. I do not, with respect, regard it as self-evident that a rider who falls off a rearing horse (or for that matter a cantering horse) is likely to suffer severe injury. It has not been suggested that the expert evidence on the likelihood of injury was inadmissible, even though Etherton LJ suggested that expert evidence need not be called. In my judgment the judge was entitled to rely on that evidence and make the finding of fact that he did. In respectful disagreement with Maurice Kay LJ I would uphold the judge’s decision on section 2 (2) (a). I have found the remaining aspects of section 2 (2) difficult, but I do not dissent from anything that Maurice Kay LJ has said about their interpretation. In particular I agree that the interpretation that the cases have placed on section 2 (2) (b) has virtually emasculated it. I also agree with what he has said about section 5 (1). On the particular facts of this case, however, I agree with Stanley Burnton LJ for the reasons that he gives that it is questionable whether section 2 (2) (b) was satisfied.

Assuming, however, that I am wrong about section 2 (2) (a) it may be that the balance can be restored by a realistic application of section 5 (2) of the Act. Voluntary acceptance of risk was not argued in Welsh v Stokes. It was in Freeman v Higher Park Farm, and it was the ground on which the keeper of the horse succeeded. To borrow again from Lord Hobhouse in Tomlinson v Congleton BC (§ 81):

“…it is not, and should never be, the policy of the law to require the protection of the foolhardy or reckless few to deprive, or interfere with, the enjoyment by the remainder of society of the liberties and amenities to which they are rightly entitled. Does the law require that all trees be cut down because some youths may climb them and fall? Does the law require the coastline and other beauty spots to be lined with warning notices? Does the law require that attractive waterside picnic spots be destroyed because of a few foolhardy individuals who choose to ignore warning notices and indulge in activities dangerous only to themselves? The answer to all these questions is, of course, no. But this is the road down which your Lordships, like other courts before, have been invited to travel and which the councils in the present case found so inviting. In truth, the arguments for the claimant have involved an attack upon the liberties of the citizen which should not be countenanced. They attack the liberty of the individual to engage in dangerous, but otherwise harmless, pastimes at his own risk and the liberty of citizens as a whole fully to enjoy the variety and quality of the landscape of this country. The pursuit of an unrestrained culture of blame and compensation has many evil consequences and one is certainly the interference with the liberty of the citizen.”

In my judgment similar considerations apply in this case. An individual who chooses to ride horses for pleasure no doubt derives enjoyment from being able to control a powerful beast. But inherent in that activity is the risk that on occasions the horse will not respond to its rider’s instructions, or will respond in a way that the rider did not intend. That is one of the risks inherent in riding horses. That is all that happened in the present case.

I too would dismiss the appeal.